Photogrammetry-Based Volume Measurement Framework for the Particle Density Estimation of LECA

Area of Science:

• Materials Science
• Geotechnical Engineering
• Computational Imaging

Background:

• Accurate particle density estimation is vital for characterizing porous aggregates like LECA.
• Traditional methods (CT, Archimedes') have limitations in precision, cost, or applicability for porous materials.
• Photogrammetry offers a potential non-destructive, cost-effective alternative for volume measurement.

Purpose of the Study:

• To present and validate a photogrammetry-based framework for LECA particle density estimation.
• To compare the accuracy of photogrammetry with Computed Tomography (CT) and Archimedes' method.
• To investigate the impact of pore interpretation on volume measurements and propose a processing method.

Main Methods:

• Development of a photogrammetry-based volume measurement framework.
• Application of an Ambient Occlusion algorithm (t=0.175) for consistent shape processing.
• Comparison of results with Computed Tomography (CT) and Archimedes' method.
• Evaluation of shape reconstruction accuracy using Hausdorff distance.

Main Results:

• The photogrammetry (Structure from Motion - SfM) method shows high accuracy, with density differences < 0.006 g/cm³ compared to CT.
• Archimedes' method tends to overestimate particle density, though acceptable for some engineering applications.
• Shape reconstruction accuracy is high, with maximum errors between 0.073 mm and 0.129 mm for 95% of the surface.

Conclusions:

• The proposed photogrammetry framework is a valid and accurate method for LECA particle density estimation.
• This approach provides valuable data for estimating intergranular porosity, essential for Discrete Element Method (DEM) model calibration.
• The methodology offers a practical tool for analyzing porous aggregates in materials science and engineering.